One
day early this August, Mitt Romney gripped a microphone at the Iowa
State Fair, faced a crowd of a few hundred people, and began, a little
joylessly and therefore a little rapidly, to give a speech. It is the
opinion of some of Romney’s friends, some of the men with whom he made
his fortune, that the repetitive business of campaigning simply bores
him and that this boredom is responsible for the fairly sizable gap
between the charismatic man they know in private and the battery-powered
figure who often appears in public.
Romney, of course, is
not the only person bored by Romney’s campaign appearances, and in the
glazed reaction of the crowds you can see some skepticism about whether a
candidacy predicated on bringing a businesslike efficiency to the
WhiteHouse—a candidacy, basically, of process—could be something to
rally around. “Over the coming decades,” Romney said at the fair, barely
pausing between each idea, “to balance our budget and not spend more
than we take in, we have to make sure that the promises we make in
Social Security, Medicaid, and Medicare are promises we can keep. And
there are various ways of doing that. One is we could raise taxes on
people …”
“Corporations!” a man cried out from
the midst of the crowd. Romney was halfway through his next sentence,
but he stopped and pivoted, noticing the hecklers, one of whom (it
turned out) was a 71-year-old former Catholic priest from Des Moines.
Morality incarnate. “Corporations!” a heckler cried again.
Romney grinned.
“Corporations are people, my friend,” he said, neatly, flatly, and
looked back to the crowd, eager to press on. Suddenly there were loud
objections coming from all over, catcalls and cries of disbelief. But
the cameras detected a splash of interest on Romney’s face.
“Of course they are,”
Romney said, and he began to explain his logic. “Everything
corporations earn ultimately goes to people. So—” Another heckler
started ostentatiously laughing, a kind of mock disbelief. The
candidate tried another tack. “Where do you think it goes?” Romney said.
“In their pockets!” someone cried out.
Romney was already a step ahead. “Whose pockets?” he said, now almost gleeful. “Whose pockets?
People’s pockets! Okay.
Human beings, my friend!”
In his quick, casual
reply—corporations are people—Romney had seemed to give something away,
though it wasn’t immediately clear what. The press chose to play the
episode as a “gaffe,” as ABC’s Jake Tapper described it, a moment in
which the weakness in Romney’s political pitch, the gap between his own
privileged experience of the world and that of working-class voters, had
been exposed. MSNBC, in a spate of giddy incredulity, seemed to keep
the clip on loop for a week. But Romney’s own campaign managers did not
try to obscure the episode at the state fair, to say he had been
misunderstood or to secret it away. Instead they promoted it, as an
advertisement of principle, and made the confrontation the centerpiece
of a solicitation to supporters. A few days later, Romney’s
communication director, Gail Gitcho, told the press that the exchange
had raised $25,000 within 24 hours.
The incident, in
retrospect, did less to peg Romney as a creature of privilege than it
did to reveal something deeper. For Romney, the corporation has long
been an object of a certain idealism. It is something he has spent much
of his adult life—first as a management-strategy consultant, then as CEO
of the private-equity firm Bain Capital—working to perfect, to strip of
its inefficiencies until it might function as a perfectly frictionless
economic unit.
The political
genuflection to businessmen is so gauzy and generic that praise for a
candidate’s private-sector acumen can often sound phony. But Mitt Romney
is the real thing. He was, by any measure, an astonishingly successful
businessman, one who spent his career explaining how business might
operate better, and who leveraged his own mind into a personal fortune
worth as much as $250 million. But much more significantly, Romney was
also a business revolutionary. Our economy went through a remarkable
shift during the eighties as Wall Street reclaimed control of American
business and sought to remake it in its own image. Romney developed one
of the tools that made this possible, pioneering the use of takeovers to
change the way a business functioned, remaking it in the name of
efficiency. “Whatever you think of his politics, you have to give him
credit,” says Steven Kaplan, a professor of finance and entrepreneurship
at the University of Chicago. “He came up with a model that was very
successful and very innovative and that now everybody uses.”
The protests going on
at Zuccotti Park now have raised the question of whether that transition
was worth it. What emerged from that long decade of change was a system
that is more productive, nimble, and efficient than the one it
replaced; it is also less equal, less stable, and more brutal. These
evolutions were not inevitable. They were the result, in part, of
particular innovations developed by a few businessmen beginning a
quarter century ago. Now one of them has a good chance of becoming
president.
|
Mitt Romney and Bill Bain, a founder of Bain & Company.
(Photo: Justine Schiavo/The Boston Globe/Getty Images)
|
Right
now, a decade into the 21st century, the character of the management
consultant is so ubiquitous a part of the global economy that McKinsey
more or less guards the gate of the modern financial world. There is a
study currently concluding in India, for instance, in which Accenture
consultants are parachuted in to village textile factories, while a
control group of factories is being kept virtually consultant-free, to
see how much the strategists can improve operations. (The early results
look good for the consultants.)
But in 1975, when Mitt
Romney graduated in the top 5 percent of his Harvard Business School
class, consulting was still a novel field. As Walter Kiechel’s
entertaining history
The Lords of Strategy documents, the three
most prestigious consultancies (then, as now, McKinsey, the Boston
Consulting Group, and Bain & Company) believed they were bringing a
newly sophisticated, quantitative approach to business, using theories
and techniques to help American industry modernize. They regarded
themselves as intellectuals, and they also paid better than anyone
else—this being a decade before Wall Street salaries started to really
climb—and so Romney made what was, at the time, an obvious career
choice. He became a consultant, first at the Boston Consulting Group and
then, three years later, at Bain & Company.
The Bain & Company
consultants who traveled the circuit of American business in the late
seventies and early eighties experienced a mass of frustrations. The
efficient, data-driven theory of business the consultancies had
developed did not in any real way cohere with the practice of business
that they saw in executive suites in St. Louis, Rochester, Houston. The
theory said that companies should focus on their core business, but
everywhere corporations were developing misguided plans to become
conglomerates. The theory said management should measure everyone’s
productivity in a firm, down to the lowliest employee, and every last
worker should be rewarded or punished depending upon his performance,
but the social relationships of business seemed to have decayed into a
long, amicable golf-course lunch. There was a loyal, almost
paternalistic attitude toward workers, protecting them even when they
seemed to be drags on growth. When I interviewed Romney’s early
colleagues about the business world that they surveyed during this
period, they tended to adopt an attitude of high disdain. “Sloppy,” one
told me. “Complacent,” said another. “Lazy,” said a third, “and out of
tune with the change that was going on in the world.”
These men are still
around—it is an unusually small and tight group, and many of them have
continued to work on and off for Romney as he has moved into public
life. They are mostly immensely rich, and if they give a collective
impression, it is tanned, engaged, upbeat, as if it is always 8 a.m. on a
Saturday and they are the fathers of shortstops. But when they began
their careers, in their own telling, they were outsiders on the make.
“If you think about that era—I grew up in the sixties and
seventies—business wasn’t a particularly noble profession,” Geoffrey
Rehnert, an early Bain partner, told me. “The best brains went into
medicine, the next best went into law.” American business, he said, “had
a thinner bench.” Those who gravitated to Bain built a culture that was
“not entitled at all. Not a single person was born with money in their
family. Every single person came from a blue-collar or middle-class
background.”
“Except for Mitt,” I
said. (Romney’s father had been the head of American Motors
Corporation, the governor of Michigan, and a member of the Nixon
cabinet; there is no credible way to describe the American elite that
excludes Mitt Romney.)
“Even he didn’t come from affluence,” Rehnert insisted. “He wasn’t a trust-fund guy.”
Perhaps what he meant
was: Romney wasn’t a Wasp. He never really talked to his co-workers
about his Mormonism, but he sometimes joked with Jewish colleagues about
how their religions made them all outsiders. Even for those who worked
with him, Romney had an inscrutable quality: They never cursed around
him and didn’t drink, and they understood that his social life would be
his family life. “I always felt that Mitt viewed himself as one of the
chosen few,” one of Romney’s colleagues at Bain Capital told me. “I
don’t think it ever affected his decision-making, but there was that
overhang.” Romney was, in the late seventies and early eighties, heavily
involved in Bain’s recruiting, and this is how many of his cohort still
view him, as a handsome guy with a great handshake. Bill Bain, the
consultancy’s titular founder, once told the New York
Times that Romney seemed a decade older than he actually was.
|
(Photo: From top, Superstock/Getty Images; Anthony Pescatore/NY Daily News Archive/Getty Images; Superstock/Getty Images)
|
Of all the business theories developing
at the time, Romney and his cohort were particularly influenced by one
that played to their sense of detachment from the business
Establishment. In 1976, two business scholars, Harvard’s Michael Jensen
and the University of Rochester’s William Meckling, published an
important paper elaborating a new idea of the firm, one that would come
to be called “agency theory.” Previous corporate theory had emphasized a
separation of powers between shareholders (who own a company) and
management (the executives who run it). This situation, Jensen and
Meckling pointed out, introduces a “principal-agent” problem, in which
each agent has incentives that run contrary to the shareholders’
interests and could hamper the firm’s ability to function.
If you were looking
across the landscape of American business 30 years ago, you could see
agency problems everywhere. In the sixties, companies had become
conglomerates so frequently that 20 percent of the Fortune 500 underwent
a merger or an acquisition in a three-year period. CEOs had enjoyed
building empires, and their shareholders, satisfied by decent returns,
had often deferred to management control. But during the stagnant
seventies, CEOs seemed loath to close factories and lay off workers. By
the early eighties, as growth once again seemed possible, shareholders
had become more restive, and innovative thinkers on Wall Street had
begun to press the case that these companies had grown inefficient and
timid, that management was underperforming.
Bain consultants did
what they could, during their assignments, to improve their clients’
operations, but they were often frustrated by an agent problem of their
own: Bain was just a consulting firm, and “a consulting firm,” says
David Dominik, an early Romney colleague, “can’t make anything happen.”
But Jensen and Meckling had sketched out one potential solution: If
managers could secure financing to run their own companies, they might
be able to build a better corporation, one that delivered stronger
returns to its owners.
You could view this
idea at least two different ways. One was as a chance to change the way
American business is run. Another was as a business opportunity to
exploit. Romney saw both.
Every
business story begins with a proposition, and the one that launched
Bain Capital was the notion that the partners might do better if they
stopped simply advising companies and starting buying and running the
firms themselves. When Bill Bain asked Romney to run the new spinoff, in
1983, the idea made sense from the perspective of Bain & Company.
The senior partners were awash in cash that they were looking to invest;
its more junior partners needed something to do. The original plan was
vague in the details, but a bowl was soon passed around the Bain &
Company boardroom so each partner could write his first name and the
amount he wanted to invest on a scrap of paper and slip it in. Romney’s
reputation was strong enough that he picked up $12 million in pledges in
that meeting alone.
Finding outside
investors wasn’t as easy. Romney went on the road, traveling to meet
with billionaire families—an investment arm of the Rockefeller fortune, a
Rothschild heir—arguing that Bain’s work in consultancy had prepared
them to turn businesses around themselves. But Romney and his cohort
were young men in their thirties with no experience investing money or
running companies, and for nearly a year the pitch kept failing. Romney
finally found some takers from Latin America, most important the
enormously wealthy Poma family, and by 1984, he and six consultants he’d
picked were staging a photo shoot for the brochure accompanying their
first fund; grinning and geeky, they posed for an outtake with dollar
bills stuffed in their mouths, their sleeves, their collars.
The leveraged-buyout
industry in its early days functioned as a laboratory for reinventing
business. Most of the promising firms were based in New York and
specialized in financial innovation—reengineering a balance sheet or
making use of new tools like junk bonds. Romney’s team in Boston looked
down on them as “just deal guys,” and at financial engineering as a
“commodity product.” Bain Capital focused instead on the way a business
runs.
Their new firm
reflected some aspects of Romney’s own personality: his mania for detail
and for process. He was a cautious executive. “Mitt was always worried
that things weren’t going to work out—he never took big risks,” one of
his colleagues told me. “Everything was very measurable. I think Mitt
had a tremendous amount of insecurity and fear of failure.” Romney never
worked from any particular “macro theme,” any philosophy of how the
economy was moving. What he employed instead was an exhausting habit of
playing devil’s advocate, proposing sequential objections to a
particular project or idea, until eventually, through a kind of
Darwinian process, consensus was reached. “I never viewed Mitt as very
decisive,” says one of his Bain Capital colleagues. “The idea was that
if there’s enough argument around an issue by bright people, ultimately
the data will prevail.” Romney may have been, as another early Bain
Capital partner puts it, a “very case-by-case, reactive thinker,” but he
was also an extremely hard worker and an egalitarian boss. He inspired
intense loyalty, and there are still members of his circle who describe
him as a perfect CEO. He was prone to profuse sweating, and the imagery
of the era is heavy on the CEO’s drenched, stained shirts.
|
Romney,
center, and six other founding members of Bain Capital, in an outtake
of their photo shoot for a 1984 brochure that surfaced this month.
|
In the mid-eighties, a
European retail outfit called Makro, a kind of continental Costco, was
looking for an executive to help run its U.S. business, and it called a
Boston supermarket executive named Tom Stemberg, inviting him to tour a
pilot store outside of Philadelphia. The store didn’t impress him much,
but he noticed that the office-supply aisle was absolutely packed with
shoppers. He told the Makro executives to abandon their model and
concentrate solely on office supplies; when they declined, he decided to
give it a try himself. Boston business is a small world, and when he
went looking for a venture-capital partner, he eventually found his way
to Mitt Romney and his new $37 million fund. “Most V.C.’s thought it
was ridiculous,” Stemberg says. “Mitt was highly unusual in that he went
to the research level to study it.”
The trouble with the
idea, to Romney’s subordinates at Bain Capital, was that the small
businesses Stemberg needed to draw weren’t accustomed to visiting a
store for office supplies; they got them from separate vendors, some who
delivered—one supplier for pens and paper, one for printer cartridges,
and so on. “Some of us were worried that we needed to change consumer
behavior,” recalls Robert F. White, one of the firm’s managing
directors. Romney persisted. As members of the group surveyed more and
more small businesses in suburban Massachusetts, they discovered that if
you asked a small-business manager how much he spent on office
supplies, he would give you a low estimate and tell you it wasn’t worth
it to send someone in a car to buy them. But if you asked the
bookkeepers, you got a far higher number, about five times as much—high
enough, Romney and Stemberg thought, to get them to come to the store.
The idea became Staples. Romney’s Bain Capital colleagues were soon
helping to select a cheaper, more efficient computer system for the
first store; they were helping stock the shelves themselves. As Staples
succeeded, and began to expand, they looked at analytics for
everything—the small-business population around a proposed store site,
traffic flow—and gamed out exactly how big a customer would need to be
before it demanded delivery. Romney sat on the Staples board for years,
and his company made nearly seven times what it invested in the
start-up.
“These
Bain Capital
guys were agents of the shareholder-value revolution.”
Romney
and his team did this sort of thing again and again, sometimes in
venture-capital deals but more often through buyouts—Brookstone,
Domino’s, Sealy, Duane Reade. In their more complex deals, they couldn’t
rely on their own team to seek out every inefficiency. They needed a
more powerful lever, and they turned to the solution Jensen and Meckling
had begun to explore a decade earlier: offering CEOs large equity
stakes in the company in the form of stock or stock options. This was a
relatively new idea, mostly untried in American business. At the same
time, a board formed in part of Bain Capital appointees who had put up
their own money in the deal would be more engaged in management details.
“You have the total alignment of incentives of ownership, board, and
management—everyone’s incentives are aligned around building shareholder
value,” Dominik says. “It really is that simple.”
In 1986, Bain Capital
bought a struggling division of Firestone that made truck wheels and
rims and renamed it Accuride. Bain took a group of managers whose
previous average income had been below $100,000 and gave them
performance incentives. This type and degree of management compensation
was also unusual, but here it led to startling results: According to an
account written by a Bain & Company fellow, the managers quickly
helped to reorganize two plants, consolidating operations—which meant,
inevitably, the shedding of unproductive labor—and when the company grew
in efficiency, these managers made $18 million in shared earnings. The
equation was simple: The men who increased the worth of the corporation
deserved a bigger and bigger percentage of its spoils. In less than two
years, when Bain Capital sold the company, it had turned an initial $5
million investment into a $121 million return.
Even by the standards
of the times, Bain Capital grew tremendously fast: from $37 million
under management in 1984 to $500 million in 1994 (and $65 billion
today). To other businesses, the buyout industry both presented a model
for better profits and posed a take-over threat. “Having the
private-equity guys out there disciplined other companies,” says Nick
Bloom, a Stanford economist. Some techniques developed in the buyout
laboratory spread. Productive workers and managers were rewarded, while
unproductive ones were cut loose. Corporations realigned themselves to
deliver more value to their shareholders, increasing dividend payments
and stock buybacks. Within a decade, ordinary businesses were giving
large stock and option packages to CEOs. Executive compensation soared.
“These Bain Capital guys,” says Neil Fligstein, an economics-sociology
professor at the University of California, Berkeley, “were agents of the
shareholder value revolution.” By the mid-nineties,
The Business Roundtable
had changed its definition of the role of a company, winnowing a broad
set of responsibilities down to a single one: increasing shareholder
value.
In
October 1994, a machine operator named Harold Kellogg gathered five of
his colleagues; borrowed a brown van from a used-car dealership in
Marion, Indiana; and began to drive east on I-90, headed for Boston,
where Romney, in his first political race, had suddenly begun to
threaten Ted Kennedy. Kellogg had worked, for eleven years, for an
office-supplies manufacturer called SCM, but a few months earlier his
plant had been acquired by a Texas-based company called American Pad and
Paper, in which Bain Capital had a majority stake. AmPad fired all of
the union workers at Kellogg’s plant, more than 250 people in total,
then hired most of them back at much lower wages; for years, they had
gotten health-care coverage as part of their pay package, but now AmPad
asked them to pay half of the costs. The whole plant walked out.
The narrative that the
Kennedy campaign had been trying to build through the summer was that
Romney was a Gordon Gekko type, but it didn’t really catch until
Kellogg and his five friends started touring Massachusetts, visiting
manufacturing plants, and then confronted Romney during an appearance at
an East Boston Columbus Day parade. Kennedy’s campaign commercials were
suddenly filled with flat midwestern accents. Romney promised, tepidly,
to meet with the Indianans, “to see if there’s anything I can do.”
Kennedy held on, and the line among political consultants was that the
Kellogg stunt had helped turn the election.
It didn’t do much to
help Kellogg. The plant in Marion closed down six months later, and the
machine operator went to work at a nearby glass company. Management sent
in Pinkerton guards and, according to a union source, took away
machinery and moved it to nonunion plants in Utah and Massachusetts.
“You had an industry where the only thing they did was converting paper
to make Siegel pads, notebooks, and copy paper,” says Marc Wolpow, who
was at the time the Bain Capital partner who worked on the AmPad deal.
Labor in the plants, he says, was nearly a commodity product—the only
thing Kellogg and his co-workers did was to move paper from one machine
to another. This could be done more cheaply at plants in China or
Indonesia. “Those jobs were going to get destroyed internationally. That
plant was going to go out of business, and there was nothing Mitt
should have done, or could have done, to prevent it.” But it is harder
to be so charitable when you look at the broader moral contours of the
arrangement. By 2001, five years after the company had been taken
public, it had filed for bankruptcy and liquidated its assets. But Bain
Capital made more than $100 million from AmPad for itself and its
investors.
After the plant
closed, the head of its union, Randy Johnson, tried to keep track of
where everyone went. He assembled a roster of the destinations of his
former colleagues—some moved to Tennessee, some to Texas—but the effort
was incomplete, and what Johnson compiled was only a partial catalogue
of loss. It’s difficult to track the fallout of any one private-equity
firm’s work, but scholars have been able to look at the conesquences of
the industry as a whole. These studies have consistently found that
private-equity takeovers improve productivity and shed jobs. But one
interesting nineties study, by two academics, Don Siegel at SUNY Stony
Brook and Frank Lichtenberg at Columbia, found something surprising:
White-collar workers, for the first time, were more vulnerable than
blue-collar workers. “Part of what the private-equity firms were doing
was replacing office workers with information technology—that’s where
they were getting some of their gains,” says Siegel, now the dean of the
University of Albany’s business school.
Here, too, private
equity seemed to provide an early warning of broader changes. In three
years during the early nineties, the Princeton economist Henry Farber
has found, roughly 10 percent of American white-collar male managers
lost their jobs. For the first time, according to data collected through
the General Social Survey, white-collar workers were nearly as worried
about losing their jobs as blue-collar workers. Those white-collar
workers who kept their jobs worked harder, and the compensation that had
once been spread through the broader middle ranks of corporations now
collected at the top. In 1980, a CEO had earned about 35 times the wages
of an average worker; by 1990, it was about 80; and by 2000, it was
about 300. The portion of America’s gross national product that ended up
in the hands of workers declined by more than 10 percent between 1979
and 1996; the portion that went to investors rose by a similar amount.
“What you end up with is a choice between a bigger cake less equally
split and a smaller cake equally split,” says Bloom, the Stanford
economist. “But that’s a social question.”
There
is no doubt that the tools of this efficiency movement helped to build
the economy of the nineties, and this fact makes Bloom’s social question
somewhat more complicated. That booming decade, with unemployment
declining by 3.5 percent and real GDP growing by nearly 4 percent each
year during the Clinton administration, depended heavily on a spike in
productivity, which itself had hinged on the wide deployment of computer
technology to displace more expensive forms of labor. Economists
believe there was a clear connection between the labor-market changes in
the early nineties and the great profits that soon followed. “Could we
have had the productivity boom without displacement? My answer would be
no,” says Frank Levy, an MIT economist.
The trouble, Levy
believes, was that this new shareholder-value-driven system had no
built-in mechanism of regulation, and its incentives geared CEOs toward
shortsightedness and recklessness. “Any profit-making organization was
going to take advantage of the opportunities to lower costs and become
more efficient by taking advantage of foreign producers and installing
technology, both of which meant losing jobs,” he says. “But
decision-makers fully exploited at every turn the market power that they
had. The question is, why were we so willing to exploit everything?”
The obvious answer is
financial reward. But there may have been a cultural component, too. By
the time Mitt Romney left Bain Capital for good, in 1999, American CEOs
looked very different from the predecessors he had met in the
seventies—the genial paternalists, spending their careers at a single
company. More and more, they were pure meritocrats—well-educated,
well-compensated, moving frequently between jobs and industries, trained
to look ruthlessly for efficiency everywhere. They look a great deal
more, in other words, like Mitt Romney.
If you trace the
public controversies over Bain Capital over time, you can see how the
obsession over shareholder value and efficiency proved not just
inequitable but destabilizing. A half-decade after Harold Kellogg
showed up in Boston, Bain Capital and others were sued by shareholders
of Stage Stores, a Texas retailer, charging Bain of helping to
manipulate the stock. The lawsuit accused the company of giving
misleadingly optimistic performance projections, which sent the
retailer’s stock soaring past $50 a share, at which point Bain Capital
unloaded virtually all of its stock. When more realistic earnings
projections were released, Stage Store’s stock plunged 58 percent in a
single day. The lawsuit was later dismissed. But then, shortly after
Romney left came the KB Toys fiasco, in which, another lawsuit alleged,
Bain Capital and KB executives took a dividend recap of over $120
million two years before the company collapsed into bankruptcy.
No court found that
Bain Capital did anything illegal in these cases. But these episodes
still give a glimpse of the evolving problems of the shareholder-value
model, and some consequences of the trade-off we made of stability for
growth. In some economic moments, a program of radical efficiency can be
good for society; at other times, when there is less fat to trim, the
same instincts can lead a company to cannibalize itself. “We’re living
in a crueler capitalism,” Fligstein says. By some measures, he adds,
“we’ve gone really quite a long ways. And nobody really knows what the
tipping point is, or how you go back.”
When
Romney was elected governor of Massachusetts in 2002, one of the
members of his transition team was Tom Stemberg, the founder of Staples.
The two men were talking one day, and Romney asked Stemberg if he had
any ideas for how he ought to govern. Stemberg, thinking off the top of
his head, had two ideas. “One was to blow up Logan airport and start
over.” That didn’t make it far. The other one did.
Stemberg served then
(as now) on the president’s council of Massachusetts General Hospital,
and he remembered a conversation he’d had with a doctor named Peter
Slavin, who has been the chief executive of that hospital system for
most of the last decade. “[Slavin] mentioned this huge problem,”
Stemberg told me, “which is all these uninsured people clog the ER.” The
hospital had to treat them. “There was a law that said that all the
insurance companies had to fund the free care. That system made
absolutely no sense. “It was, as Stemberg told Romney, “the least
efficient way to serve them.” The conversation moved on, and Stemberg
figured his ten-minute-long career as a policy maven was over. “I
figured that’s the end of that.” But Romney’s staffers, consulting with
experts, began to work out a fix, requiring almost every citizen in the
state to carry insurance, and providing subsidies for those who couldn’t
afford it. Eventually he was heading down to Ted Kennedy’s office in
Washington to explain the program, PowerPoints in hand. Three and a
half years later, Romney introduced his universal-health-care plan, and
in the press he credited Stemberg with suggesting it.
The
punch line, of course—the punch line to Romney’s campaign so far—is
that the plan he built was an almost exact model for Obama’s national
plan, designed by some of the same experts.
But what separates
Romney’s plan from Obama’s—and gives some clues about his potential
presidency—is its almost-accidental origin. Romney did not begin with a
philosophical quest to improve American health care. He began with the
idea of himself as a problem solver and asked those around him for a
problem that he might usefully solve. I remembered, when I was told this
story, an anecdote I’d heard from a former political staffer of
Romney’s. On even basic philosophical questions like abortion, the
staffer said, Romney did not try to resolve the question in the
abstract, as a matter of principle, and would consider instead various
hypothetical cases—for instance, a late-term abortion—and build from
them a politics. The line that Romney is a flip-flopper may vastly
understate the depth of the condition.
It is arresting to
imagine a Romney White House, inevitably filled with as many former Bain
colleagues as each of his other public ventures have been: The
PowerPoints, the 80-20 jargon, the clinical separation of
decision-making from ideology, the detachment of those decisions from
moral consequence, a persistent blind spot for people as people. It
would represent the final ascension of a perfectly American type, one
that has already remade the culture of business. I once asked a Bain
colleague of Romney’s how Romney thought of his own core competence. “I
think Mitt thinks he’s good at being Mitt Romney,” the colleague said.
But Romney’s
career-long commitment to his own particular brand of impersonal
decision-making might suggest something personal after all. One great
mystery about Romney has been where his Mormonism comes in and what it
explains. Maybe the clearest answer comes from taking at their word the
businessmen with whom he came up, who say they never saw its influence.
Romney’s religion constitutes a minority set of beliefs. Poorly
understood and widely mocked, it can provoke suspicions about his
motives. Perhaps it is not surprising, then, that he has adopted a
public persona that contains no detectable motives at all, one that is
buried in objectivity, in data, in process. The best evidence of how
important Romney’s religion is to him could be how far he has kept it
from view. But the character that remains visible is at once uniquely
American and a little strange: a perfectly objective efficiency machine.
No comments:
Post a Comment